Dallas Keuchel: The Unheralded Reason the White Sox are in 1st

On the heels of Lucas Giolito’s no-hitter, the unspectacular happened. Dallas Keuchel toed the rubber and gave the White Sox a solid six inning performance. It was his seventh outing this year and his seventh time pitching 5.1+ innings allowing three runs or less. He stands as the rock of the rotation; the robin to Giolito’s batman.

The White Sox signed him this offseason to do just that. It was a 3 year, $55 million investment to be a workhorse and provide stability to a rotation of budding pitching prospects. However, while many teams would have paid Keuchel for what he was, the White Sox had a plan to make him better.

For the majority of his career, Dallas Keuchel has been a sinker-slider pitcher. This year has seen him decrease usage of his sinker, increase cutter usage, and nearly double changeup usage. He’ll still mix in four-seamers and sliders in two strike counts, but not nearly to the extent he used to.

He has effectively shifted from being a five pitch pitcher in 2018 to a three pitch pitcher in 2020. Keuchel hinted at some of these changes in 2019. He threw a career low rate of four-seamers, career high rate of cutters, and his lowest rate of sliders since his rookie season. On top of accentuating the 2019 trends, the White Sox had their own twists to add.

The increase changeup usage was purely due to throwing it more versus right handed batters – he has actually reduced its usage against lefties. About one third of the pitches Dallas has thrown to righties this year have been changeups, up from a career rate around 15%.

His approach against left handed batters has seen an even more drastic change. After years of not throwing cutters to lefties, it has suddenly become an integral part of his attack plan. From 2017 to 2019, cutters made up less than one percent of Keuchel’s pitches to lefties; that has risen to 31.7% in 2020. Both of these strategy changes have largely been offset by the decreased sinker usage.

This was not simply a case of one pitch being revamped and accordingly having its usage increased; this is a wholly rethought approach to batters of both handedness. The White Sox acquired a player who had already found success in the major leagues and were willing to take a chance to make him even better.

Thus far, just looking at the surface statistics, the experiment seems to be a success. Through seven starts he is 5-2 with a 2.70 ERA in 43.1 innings pitched. Then again, surface stats can lie. Dylan Cease currently has a 3.00 ERA despite possessing a horrible strike-to-walk ratio (K:BB) and a hazardous home run habit. His stats may look fine thus far, but his peripheral statistics suggest he’s a ticking time bomb. What do Keuchel’s peripheral stats have to say? Has he rediscovered his All-Star-self or is he, like his teammate Dylan Cease, due for regression?

Both his walk and strikeout rates have gone down. His K:BB of 2.5 is around where he’s stood the last few years, but the general decrease makes things interesting. Fewer walks/strikeouts means he’s allowing more balls in play leaving him subject to high volatility. It makes the softness of contact he’s inducing crucial. If batters are crushing baseballs, there’s little Keuchel can do to slow their efforts.

Of all the teams to be allowing batted balls in play on, the White Sox are probably the best. They lead the MLB with a team Defensive Runs Saved (DRS) of 28 meaning, relative to a league average defense, their defense has already saved about 28 runs this season. They generally make balls in play more forgiving, taking away hits that may otherwise fall.

Keuchel’s batting average on balls in play (BABIP) has been a suspiciously low .244. With a large enough sample size to regress, most players sit around .300 and given his career .294 BABIP, he does not seem to be an exception. The top defensive units last year were able to sustain team BABIPs of just over .270, but that appears to be the realistic cap. The difference between his wOBA and xwOBA similarly suggest he has been lucky, even considering the elite defense behind him. His difference (.141) is triple the gap the White Sox defense has been able to sustain teamwide (.047). 

Keuchel’s route to salvation is through home run prevention. He’s led the AL in home run rate among qualified starters thus far into the season. He may not be able to sustain this low a rate, but looking at previous years, he just might be able to sustain a rate around 0.7-0.9 home runs per nine innings (HR/9). While his expected wOBA on contact (xwOBAcon) has been high, he’s kept exit velocities low in 2020. When Keuchel has kept his exit velocities low in the statcast era, it has translated to low xwOBAcons and impressive home run rates. 

FIP – an ERA predictor based on walks, strikeouts, and home runs – puts a lot of weight on the ability to prevent home runs. This makes sense given how quickly they can turn a game on its head. Considering this, it will likely be the difference between Keuchel being good and mediocre. This is often the case for pitchers, but this is a case of statcast metrics disagreeing. His xwOBAcon suggests he will be allowing a near league average home run rate (as in 2016 and 2019) while his exit velocity suggests he should be able to prevent at a level to keep his ERA below 4.00 (as in 2015, 2017, and 2018). 

The White Sox are in dire need of starting pitching. With Gio Gonzalez and Carlos Rodon on the IL and the FIP of Dylan Cease and Reynaldo Lopez being over 6, it makes sense they’re looking to acquire a big arm before the trade deadline. As much as they could use a Lance Lynn or Mike Clevinger, they need Keuchel to remain a rock in the rotation. Ironically, as Keuchel moves away from his sinker in the most home run happy era in baseball history, he needs to keep doing what sinkerballers have done for decades: prevent home runs.